The impact of immortal time bias on the estimated mortality benefit of antifibrotic therapy in IPF

<bold>Background:</bold> The survival benefit of antifibrotic therapy in idiopathic pulmonary fibrosis (IPF) remains uncertain. The presence of immortal time bias in observational studies can overestimate the mortality benefit. This occurs where patients are classified as “treated” from the start of observation, instead of being reclassified at start of treatment, creating a period of immortality. <bold>Objective:</bold> Assess the mortality benefit of antifibrotic therapy while accounting for immortal time bias. <bold>Methods:</bold> IPF patients (N=969) from Nordic registries were included in a retrospective cohort study. The time-dependent effect of antifibrotic treatment on transplant-free survival (TFS) was assessed using Kaplan-Meier estimates and Cox proportional hazards models. <bold>Results:</bold> Of 969 patients, 803 (83%) received nintedanib or pirfenidone. Treatment was associated with lower mortality when assessed as a time-dependent variable, with median TFS 4.6 years on treatment and 3.6 years without treatment (p=0.006). Treatment was not associated with survival benefit after adjustment for age, FVC and DLCO. By contrast, median TFS appeared to be 4.7 years among treated and 2.2 years among untreated patients when not accounting for immortal time bias (Figure). <bold>Conclusion:</bold> Immortal time bias appears to inflate the apparent mortality benefit of antifibrotic therapy in IPF, and should be accounted for in observational studies. <fig><object-id>erj;64/suppl_68/PA2433/F1</object-id><object-id>F1</object-id><object-id>F1</object-id><graphic></graphic></fig>